NEW ORLEANS, May 18 (UPI) -- Google Flu Trends influenza surveillance is cheap and timely but should be interpreted with caution, U.S. researchers said.
Dr. Justin Ortiz, a clinical fellow at the University of Washington, said the previous Google Flu Trends validation study showed the surveillance for influenza-like illness was somewhat accurate. However, the current study of Google Flu Trends found it was not as accurate as the Centers for Disease Control and Prevention's rates of laboratory-confirmed influenza.
Google Flu Trends uses certain Google search queries in real time to estimate nationwide rates of influenza-like illness activity involving symptoms of fever, cough or sore throat, without any laboratory testing.
The researchers compared Google Flu Trends data to CDC's national surveillance and found that Google Flu Trends was 25 percent less accurate at estimating rates of laboratory confirmed influenza virus infection.
There are many influenza-like illnesses that are not influenza -- in fact, studies show that during flu season, 20 percent to 70 percent of cases are caused by flu, Ortiz said.
"There is a wide and unpredictable proportion of influenza-like illness due to the influenza virus," Ortiz said in a statement. "Google Flu Trends estimates of influenza-like illness may not necessarily correlate with actual influenza virus infections."
The study is scheduled to be presented at the American Thoracic Society International Conference in New Orleans.